Write a 12 to 20 message story with a twist, render it as an iMessage chat with typing indicators, and export in 9:16. Full workflow plus 20 ideas.
To make a fake text message video you write a short conversation with a twist, render it as an iMessage-style chat that reveals one bubble at a time with typing indicators, add a voice or a sound bed, and export it in 9:16. A generator does the chat, the timing and the captions in one pass.
The format borrows a habit people already have. A viewer reads a text thread faster than they process a talking head, and they read it without deciding to. The video is over before the decision to scroll gets made.
Three mechanics do the work:
It is also the cheapest faceless format to produce. There is no b-roll to find and no visuals to art direct. The entire video is words on a screen with timing.
Twelve to twenty messages is roughly 15 to 30 seconds at a natural reveal speed. That is the sweet spot for TikTok and Reels. Write to that budget from the start instead of cutting later.
Use a four-beat structure and assign a message count to each beat.
| Beat | Messages | Job |
|---|---|---|
| Normal | 2 to 3 | Establish who is talking and what they want |
| Wrong note | 3 to 5 | One detail that does not fit |
| Escalation | 4 to 7 | The other person notices, and presses |
| Twist | 2 to 3 | The reveal, then silence or one last line |
Rules that matter more than the structure:
The format is the craft. A well-written story with wrong formatting still reads as fake, and a viewer who notices the interface stops following the story.
Who speaks where. The narrator, meaning the person the viewer is meant to identify with, goes on the right in blue. The other person goes left in gray. Do not swap sides mid-video.
The contact name. A name at the top does more work than any caption: "Mom", "Unknown Number", "Do Not Answer", "Landlord". Half your hook can live in that one field.
Typing indicators. Put the three dots before any reply that carries weight. Hold them slightly longer than feels comfortable. That pause is the tension.
Reveal timing. Short messages appear fast, long ones sit longer. As a rough rule, give each bubble about 0.6 to 1.2 seconds plus reading time, and give the twist a full beat of stillness before the video ends.
Read time. Never let a bubble leave the screen before it can be read twice at phone distance. Viewers rewatch text videos, and a message they missed is the reason they rewatch, not a reason they leave.
The safe area. Keep the last few bubbles out of the bottom 15% of the frame. That strip is covered by the platform interface, and a twist hidden behind a caption box is a wasted video.
These are two different formats with the same script, and they behave differently. Pick per story rather than per channel.
| Full-screen iMessage | Text message over gameplay | |
|---|---|---|
| Best length | 10 to 30 seconds | 30 to 90 seconds |
| Audio | Sound bed, notification sounds | AI voice reading the chat |
| Reading load | The viewer reads everything | The voice carries it |
| Best stories | Fast twists, one-line punchlines | Longer arcs, slow reveals |
| Feels like | Looking at someone's phone | Watching a story |
The full-screen version lives or dies on the writing, because there is nothing else on screen. The gameplay version is more forgiving: the motion underneath holds the eye while the voice reads, so a story can take 60 seconds without sagging.
Two tools cover the two versions. The fake text message video generator produces the chat over a gameplay background with a voice reading it. The iMessage video generator produces the full-screen conversation with typing animations. Same script, two exports, two different audiences.
The manual route means building a chat layout, animating each bubble, faking the typing dots, and timing all of it by hand. The generator route is one screen: you give it the conversation or the premise, pick the format, and the chat, the reveal timing, the voice and the background are produced together as a 9:16 file.
A text conversation plays out bubble by bubble while gameplay runs behind it. Two AI voices read the thread, so it works with the sound on or off.
MakeViral runs in the browser, so there is no editing software to install, and it takes a prompt, a script or a product URL as the input. Bulk generation is the part that matters for a channel: a batch of ten premises becomes ten finished videos in one sitting, which is what daily posting actually requires. Pricing as of September 2026 is $49.99 a month for 500 credits or $199.99 a month for 3,000 credits on the pricing page.
Worth knowing, because sometimes you want a layout no template offers. This is the outline in a free editor such as CapCut.
Budget 30 to 60 minutes for the first one. The repetitive part is the bubble keyframes, and it is the reason most people who make this format daily stop doing it by hand after the first week.
Five genres, four premises each. Write them in your own words and invent every name.
Relationship twists
Creepy and unsettling
Money and scams
Family drama
Work and revenge
Each premise produces a second video: the same conversation from the other person's phone. That reversal is a whole content series and it costs one rewrite.
The format earns the same way every faceless format does, and the earning happens at the channel level rather than the video level.
Creator Rewards pays on qualified views only, so enter the qualified number if you have it.
Creator Rewards reports commonly land between $0.40 and $1.00 per 1,000 qualified views. Only views over 5 seconds on videos longer than one minute qualify, so your qualified count is lower than your view count.
For the numbers behind these programs, see how to make Reddit story videos, which covers the same payout mechanics with a longer format, and the TikTok Creator Rewards Program guide.
Nine failures account for almost every flat video in this format.
Text stories are one of eight formats built on the same production stack. The faceless video generator hub lists all of them. For the longer narrative version of this workflow, read how to make Reddit story videos. For the gameplay-background style these videos often sit inside, read how to make brainrot videos. To build the channel that carries them, start with how to start a faceless YouTube channel.
A text conversation plays out bubble by bubble while gameplay runs behind it. Two AI voices read the thread, so it works with the sound on or off.
Read moreToolA whole-screen iPhone conversation: the keyboard types, the dots appear, the chime lands, the read receipt shows. No phone and no screen recording.
Read moreToolGive it a premise, get a finished vertical video: Reddit post card, AI voice, word-by-word captions and gameplay behind it. The format that still holds attention for 90 seconds.
Read morePick a story, rewrite it for voice, add an AI voiceover, captions and a gameplay background. The full 2026 workflow, with 20 story ideas. Start now.
Read moreArticleGameplay, an AI voice and hard captions, stacked in 9:16. The scripts, the backgrounds, the copyright rules and the posting cadence. Make one now.
Read moreGuideWhat a faceless YouTube channel is, the 2026 monetization thresholds, 12 niches with reported RPM, and a 30-day launch plan. Start your channel today.
Read moreBack to the pillar guide: Faceless YouTube Channel: How to Start One in 2026See every faceless video generator
Pick a format, paste a prompt or a product URL, and download a 9:16 video that is ready to post.
Cancel anytime. 14-day refund window on unused credits.